The burden of active infection and anti-SARS-CoV-2 IgG antibodies in the general population: Results from a statewide survey in Karnataka, India (preprint)

ABSTRACT

BackgroundGlobally, the routinely used case-based reporting and IgG serosurveys underestimate the actual prevalence of COVID-19. Simultaneous estimation of IgG antibodies and active SARS-CoV-2 markers can provide a more accurate estimation. MethodsA cross-sectional survey of 16416 people covering all risk groups was done between 3-16 September 2020 using the state of Karnatakas infrastructure of 290 hospitals across all 30 districts. All participants were subjected to simultaneous detection of SARS-CoV-2 IgG using a commercial ELISA kit, SARS-CoV-2 antigen using a rapid antigen detection test (RAT), and reverse transcription-polymerase chain reaction (RT-PCR) for RNA detection. Maximum-likelihood estimation was used for joint estimation of the adjusted IgG, active, and total prevalence, while multinomial regression identified predictors. FindingsThe overall adjusted prevalence of COVID-19 in Karnataka was 27 {middle dot}3% (95% CI 25 {middle dot}7-28 {middle dot}9), including IgG 16 {middle dot}4% (95% CI 15 {middle dot}1 - 17 {middle dot}7) and active infection 12 {middle dot}7% (95% CI 11 {middle dot}5-13 {middle dot}9). The case-to-infection ratio was 140, and the infection fatality rate was 0 {middle dot}05%. Influenza-like symptoms or contact with a COVID-19 positive patient are good predictors of active infection. The RAT kits had higher sensitivity (68%) in symptomatic participants compared to 47% in asymptomatic. InterpretationThis is the first comprehensive survey providing accurate estimates of the COVID-19 burden anywhere in the world. Further, our findings provide a reasonable approximation of population immunity threshold levels. Using the RAT kits and following the syndromic approach can be useful in screening and monitoring COVID-19. Leveraging existing surveillance platforms, coupled with appropriate methods and sampling framework, renders our model replicable in other settings.